Interpolation of video compression frames
Coding techniques for a video image compression system involve improving an image quality of a sequence of two or more bi-directionally predicted intermediate frames, where each of the frames includes multiple pixels. One method involves determining a brightness value of at least one pixel of each bi-directionally predicted intermediate frame in the sequence as an equal average of brightness values of pixels in non-bidirectionally predicted frames bracketing the sequence of bi-directionally predicted intermediate frames. The brightness values of the pixels in at least one of the non-bidirectionally predicted frames is converted from a non-linear representation.
1. An apparatus comprising a video bitstream stored on one or more non-transitory machine-readable media, the video bitstream characterized by:
data representing predicted and bidirectional predicted picture frames that include picture regions in a compressed format, wherein a portion of the data that represent at least one picture region in the compressed format within a bidirectional predicted picture frame comprises:
a signal indicating whether a bidirectional prediction mode for the at least one picture region is used;
a signal indicating whether an interpolative motion vector prediction mode for the at least one picture region based on two referenceable frames in the video bitstream is used;
a signal indicating whether a pixel interpolation mode using pixel weighting from the two or more referenceable frames in the video bitstream is used; and
two pixel weight values for the pixel weighting, wherein the two pixel weight values are unequal.
2. The apparatus of claim 1 , wherein the pixel weight values comprise weights in a range of −1.0 to +2.0.
3. The apparatus of claim 1 , wherein the pixel weight values comprise at least one weight greater than +2.0.
4. An apparatus to compress a sequence of picture frames, the apparatus comprising:
an input to receive picture frames in a sequence of frames; and
one or more processors, wherein the one or more processors execute instructions to:
access a picture region within a picture frame in the sequence of frames; and
encode pixel values of the picture region to generate a coded bitstream, wherein the encoding comprises:
determining pixel interpolation weights to interpolate the pixel values of the picture region using pixel values from two referenceable frames, wherein the pixel interpolation weights are unequal;
encoding, into the coded bitstream, the pixel interpolation weights;
encoding, into the coded bitstream, a signal indicating that a bidirectional prediction mode for the picture region is used; and
encoding, into the coded bitstream, a signal indicating that an interpolative motion vector prediction mode for the picture region based on the two referenceable frames in the bitstream is used.
5. The apparatus of claim 4 , wherein the pixel interpolation weights comprise weights in a range of −1.0 to +2.0.
6. The apparatus of claim 4 , wherein the pixel interpolation weights comprise a least one weight greater than +2.0.